The association between alcohol consumption and
bladder cancer risk has been insufficiently investigated in East Asian populations, who frequently have the inactive
enzyme for metabolizing
acetaldehyde. Given that
acetaldehyde associated with alcohol consumption is assessed as a
carcinogen, consideration of differences in
acetaldehyde exposure would aid accuracy in assessing the
bladder cancer risk associated with alcohol consumption. Here, we conducted a population-based cohort study in Japan to examine this association, including information on the
flushing response as a
surrogate marker of the capacity of
acetaldehyde metabolism. Hazard ratios (HR) and 95% confidence intervals (CI) were estimated using multivariate Cox proportional hazard models. During follow up from 1990 through 2012 for the 95,915 subjects (45,649 men and 50,266 women, aged 40-69 years), 354 men and 110 women were newly diagnosed with
bladder cancer. No significant association between alcohol consumption and
bladder cancer risk was observed in the overall analysis. Among male flushers, HRs were 1.04 (95% CI 0.70-1.54), 1.67 (1.16-2.42), 1.02 (0.62-1.67) and 0.63 (0.33-1.20) for alcohol consumption of 1-150, 151-300, 301-450, >450 g/week of pure
ethanol compared with non-drinkers and occasional drinkers, respectively, indicating an inverted U-shaped association between alcohol consumption and
bladder cancer risk. In contrast, no significant association was identified among male non-flushers. The marginally significant interaction between alcohol consumption and the
flushing response (p for interaction = 0.083) may support our hypothesis that
acetaldehyde derived from alcohol consumption is associated with
bladder cancer risk. A prospective study considering polymorphisms of genes involved in
acetaldehyde metabolism is warranted.